Article Text

Download PDFPDF
OC-060 Analysis of unexpected non-colorectal lesional histopathology in 1252 resections for suspected crlm
  1. D O’Reilly,
  2. J Hodson,
  3. R Marudanayagam,
  4. R Sutcliffe,
  5. P Muiesan,
  6. D Mirza,
  7. J Isaac,
  8. K Roberts
  1. Liver Unit, QE Birmingham, Birmingham, UK


Introduction Resection is a safe and effective treatment for colorectal liver metastases (CRLM). Surgery is indicated by imaging findings with a history of synchronous or metachronous colorectal cancer. This study examined cases where patients were found to have non-colorectal lesions after resection for CRLM.

Method Liver resections were identified from a prospective database. Additional data were retrieved from electronic records. Resections were divided into those where histology demonstrated CLRM or complete pathological response to chemotherapy (A) and those where an alternative pathology was detected (B). Univariate analyses with Mann-Whitney, Fisher’s exact and Kendall’s tau tests were performed.

Results There were 1252 resections (2002–2012). Group A 1213 (96.9%), Group B 39 (3.1%). The groups did not differ in age, sex or synchronous/metachronous disease. Group B patients had significantly fewer tumours (p = 0.006), with significantly smaller maximum sizes (p < 0.001). Group B patients also had significantly lower T (p = 0.031), N (p < 0.001) and Dukes’ (p < 0.001) stages. In addition to this, the CEA level was also significantly lower in group B patients (median A = 8.1, B = 1.8, p < 0.001). Group B histopathology is shown in the Table 1.

Abstract OC-060 Table 1

25/31 in group B had preoperative MRI (data unavailable in 8). 32 resections were undertaken where the primary is known to have been T0N0 (Dukes’ A). Of these 5 (16%) were in group B; two of these had malignant pathology (HCC and cholangiocarcinoma).

Conclusion Unexpected non-colorectal lesional pathology findings after resection for CRLM are rare, of the same order of frequency as mortality. In a small proportion of the cases there was some equivocation in the reports of preoperative imaging. However, all proceed on the basis they were CRLM. Haemangiomata, which were often hyalinised, were the most common finding. 9/39 had alternative malignant pathology. Current practice is to avoid lesional biopsy before liver resection because of the risk of tumour seeding. The present data support this.

Disclosure of interest None Declared.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.